Recent studies provide strong evidence that single myosin class V molecules transport vesicles and organelles processively along F-actin, taking several 36-nm steps, 'hand over hand', for each diffusional encounter. The mechanisms regulating myosin-V's processivity remain unknown. Here, we have used an optical-tweezers-based transducer to measure the effect of load on the mechanical interactions between rabbit skeletal F-actin and a single head of mouse brain myosin-V, which produces its working stroke in two phases. We found that the lifetimes of the first phase of the working stroke changed exponentially and about 10-fold over a range of pushing and pulling forces of +/- 1.5 pN. Stiffness measurements suggest that intramolecular forces could approach 3.6 pN when both heads are bound to F-actin, in which case, extrapolation would predict the detachment kinetics of the front head to slow down 50-fold and the kinetics of the rear head to accelerate, respectively. This synchronizing effect on the chemo-mechanical cycles of the heads increases the probability of the trail head detaching first and causes a strong increase in the number of forward steps per diffusional encounter over a system with no strain dependence. Myosin V has an extremely long neck containing six calmodulin-binding IQ motifs that allows it to take multiple 36 nm steps corresponding to the pseudo-repeat of actin. To further investigate how myosin V moves processively on actin filaments, we altered the length of the neck by adding or deleting IQ motifs in myosin constructs lacking the globular tail domain. These myosin V IQ mutants were fluorescently labeled by exchange of a single Cy3-labeled calmodulin into the neck region of one head. We measured the step-size of these individual IQ mutants with nanometer precision and sub-second resolution using FIONA. The step-size was proportional to neck length for constructs containing 2, 4, 6 and 8 IQ motifs, providing strong support for the swinging lever arm model of myosin motility. In addition, the kinetics of stepping provided additional support for the hand-over-hand model whereby the two heads alternately assume the leading position. Interestingly, the 8 IQ myosin V mutants gave a broad distribution of step-sizes with multiple peaks, suggesting that this mutant has many choices of binding sites on an actin filament. These data demonstrate that the step-size of myosin V is affected by the length of its neck and is not solely determined by the pseudo-repeat of the actin filament.